Iron oxide-impregnated paper vs. Bray-1 soil-test methods predicting crop response from phosphate-rock sources

The effectiveness of a Pi (iron oxide-impregnated paper) method was compared with that of Bray-1 method in extracting phosphorus (P) from an acid Hartsells silt loam (Typic Hapludult, pH 4.8) treated with low-release P fertilizers derived from Dorowa (Zimbabwe) phosphate rock (DPR) with maize (Zea mays L) as indicator crop in a randomized block design greenhouse experiment with three replicates. The fertilizers used were finely ground DPR, partially acidulated DPR (PADPR), a compacted mixture of DPR + triple superphosphate (TSP) + urea + potassium chloride (DTUK) with half of P from DPR and half from TSP, and a commercial grade single superphosphate (SSP) as a standard. The results indicated that the correlation between Bray-1 and Pi-extractable P was high (r=97***). The regression coefficients for dry matter yield (DMY) and P uptake (PUPT) as a function of Bray-1 P level of SSP (DMY = 1.16 + 0.24Bray-1 P; PUPT = 0.93 + 0.45Bray-1 P) were significantly lower (p < 0.001) than those for DTUK (DMY = 1.16 + 0.43Bray-1 P; PUPT = 0.93 + 0.74Bray-1 P) and PADPR (DMY = 1.16 + 0.47Bray-1 P; PUPT = 0.93 + 0.82Bray-1 P). That is, a given soil-extractable P level with SSP resulted in a lower dry matter yield or P uptake than that from DTUK and PADPR. Thus, Bray-1 underestimated the agronomic effectiveness of DTUK and PADPR with respect to SSP. Based on this green-house research, soil testing for fertilizer recommendations using the Bray-1 test may result in two different response functions, one for water-soluble P sources and another for water-insoluble P sources.